Update README.md
Browse files
README.md
CHANGED
@@ -8,8 +8,25 @@ datasets:
|
|
8 |
---
|
9 |
## Model description
|
10 |
Anime face generator model using [TensorFlow DCGAN example](https://www.tensorflow.org/tutorials/generative/dcgan).
|
|
|
11 |
## Training and evaluation data
|
12 |
-
Model is trained on [anime faces dataset](https://huggingface.co/datasets/merve/anime-faces).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
## Intended use and biases
|
14 |
This model is not intended for production.
|
15 |
|
|
|
8 |
---
|
9 |
## Model description
|
10 |
Anime face generator model using [TensorFlow DCGAN example](https://www.tensorflow.org/tutorials/generative/dcgan).
|
11 |
+
|
12 |
## Training and evaluation data
|
13 |
+
Model is trained on [anime faces dataset](https://huggingface.co/datasets/merve/anime-faces). The dataset consists of 21551 anime faces scraped from www.getchu.com, which are then cropped using the anime face detection algorithm [here](https://github.com/nagadomi/lbpcascade_animeface). All images are resized to 64 * 64 for the sake of convenience. The model takes a noise as input and then Conv2DTranspose is used to do upsampling. If you want to pass this to another discriminator, the output shape consists of 28x28 images.
|
14 |
+
|
15 |
+
## How to use this model
|
16 |
+
You can use this model to generate new anime faces. If you want to continuously train, use with [discriminator](https://huggingface.co/merve/anime-faces-discriminator) using `tf.GradientTape()` as mentioned in the DCGAN tutorial.
|
17 |
+
|
18 |
+
```
|
19 |
+
from huggingface_hub import from_pretrained_keras
|
20 |
+
|
21 |
+
model = from_pretrained_keras("merve/anime-faces-generator")
|
22 |
+
```
|
23 |
+
|
24 |
+
You can generate examples using a noise.
|
25 |
+
|
26 |
+
```
|
27 |
+
seed = tf.random.normal([number_of_examples_to_generate, noise])
|
28 |
+
predictions = model(seed, training=False) # inference mode
|
29 |
+
```
|
30 |
## Intended use and biases
|
31 |
This model is not intended for production.
|
32 |
|